One-Class Classifier Based Fault Detection in Distribution Systems With Varying Penetration Levels of Distributed Energy Resources

Zhidi Lin,Xuemin Hong,Liuqing Yang,Shuguang Cui,Dongliang Duan,Xiang Cheng,Qi Yang

doi:10.1109/access.2020.3009385

Abstract

The integration of Distributed Energy Resources (DERs) into distribution systems greatly increases the system complexity and introduces two-way power flow. Conventional protection schemes are based upon local measurements and simple linear system models, and are thus not capable of handling the new complexity and power flow patterns in systems with high DER penetration. In this paper, we propose a data-driven protection framework to address the challenges introduced by DERs. Firstly, considering the limited available data under fault conditions, we adopt the Support Vector Data Description (SVDD) method, a commonly used one-class classifier, for distribution system fault detection, which only requires the normal data for its training process. Secondly, incremental learning is incorporated into the proposed SVDD-based protection framework to accommodate variations of the integration level of DERs in distribution systems over time. In particular, the artificial uniform-hyperspherical data generation model is incorporated into the incremental SVDD to boost the training speed. Finally, we validate the proposed method under the IEEE 123-node test feeder. Simulation results demonstrate that our proposed SVDD-based fault detection framework significantly improves the robustness and resilience against DERs in comparison with conventional protection systems. Meanwhile, the proposed online updating model outperforms the existing incremental SVDD models in terms of successful training speed.

Highlights

We propose a data-driven fault detection framework with the following features: 1) We utilize the global data with more measurements for fault detection, as opposed to the local measurement used in conventional relay operations; 2) We adopt the one-class classification method using only the normal data as the training data, in contrast to most existing data-driven methods that rely on comprehensive fault data; 3) We take advantage of the incremental approach to meet the online updating challenge and adapt to the dynamic environment, instead of being limited to offline training for existing data-driven methods
WORK In this paper, a novel fault detection framework based on the Support Vector Data Description (SVDD) algorithm was proposed for distribution systems with varying Distributed Energy Resources (DERs) penetration
While the traditional protection usually sets a fixed threshold for all relays and fails to adapt to the changes introduced by DERs, the SVDDbased method can describe a large variety of operating conditions and conduct online model update via incremental learning

Summary

Introduction

A. BACKGROUND With the increasing integration of renewable energy resources into the power grid, the topology of distribution systems has. Conventional protection schemes are designed under the assumption of one-way power flow and the operations of protection devices are based upon only local measurements, which fail in systems with high DER. Z. Lin et al.: One-Class Classifier Based Fault Detection in Distribution Systems With Varying Penetration Levels of DERs penetration [1]. The system is highly dynamic with varying levels of DER penetration over time, which brings even bigger challenges to the traditional schemes. An effective fault protection system that can address the system complexity and dynamics brought by DERs to ensure the reliability in power systems becomes an urgent issue for the successful integration of renewable energy into the distribution system

Objectives

Methods

Findings

Conclusion